[Insight-users] What is GetNeighborhoodIndex of the ShapedNeighborhoodIterator?

[Dawood AlMasslawi]

>
> Yes, that makes sense!
>
> In the past I also had some trouble with neighborhood iterators  of a
>
> large radius, so I guess they are not utilized for that use. Probably
>
> it has something to do with the boundary condition, every time I
>
> iterated near the boundaries somethings went wrong!
>
> Dawood
>
>
>
> On Mon, Jul 18, 2011 at 9:02 PM, David Doria <daviddoria at gmail.com> wrote:
>
>> On Mon, Jul 18, 2011 at 10:35 AM, Dawood AlMasslawi <masslawi at gmail.com>wrote:
>>
>>> Following the same logic it should return the neighborhood index of
>>>>
>>>> the pixel which the constant iterator is currently accessing.
>>>>
>>>> In the wiki example since the radius of the neighborhood is set
>>>>
>>>> equal to the size of the image I guess it returns the number of the
>>>> pixel
>>>>
>>>> as the iterator walks from the top left to the bottom right.
>>>>
>>>> Hence the neighborhood index would start with 0 and end with
>>>>
>>>> number of pixels minus 1.
>>>>
>>>> Still, 199 and 241 seems a bit much, the image has only 100 pixels!!
>>>>
>>>>
>>>> Dawood
>>>>
>>>
>> I see. I changed it to:
>>
>> itk::Size<2> radius;
>> radius.Fill(1);
>> IteratorType iterator(radius, image, image->GetLargestPossibleRegion());
>>
>> and now the output is:
>>
>> New position:
>> At [0, 0]
>> 1 -> [0, -1] = 0
>> At [0, 0]
>> 7 -> [0, 1] = 0
>>
>> Which makes sense, because I set the (0,1) and (0,-1) pixels as active, so
>> if the neighborhood positions are indexed as:
>>
>> 0 1 2
>> 3 4 5
>> 6 7 8
>>
>> that makes sense. I don't know what the issue was with the larger radius -
>> maybe it just isn't designed to have a radius on the order of the size of
>> the image?
>>
>> David
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.itk.org/pipermail/insight-users/attachments/20110718/eec0117b/attachment.htm>